The broad aim of the proposed study is to improve medication adherence in adolescent kidney transplant recipients. The specific aims are to determine, in a randomized clinical trial, the efficacy of a structured, multi-component intervention in improving adherence to anti-rejection medications and graft outcomes, and to identify characteristics of healthcare systems that are independently associated with adherence. Medication non-adherence is a major problem in the 15-30% of American children who have a chronic illness resulting in significant morbidity, mortality and related costs. An intervention to improve adherence in the kidney transplant population would improve survival and quality of life in this population, and may be applicable to other populations. Kidney transplant recipients at 7 pediatric transplant centers in the United States and Canada will be invited to participate. Participants will be randomly assigned to either the control or intervention group. Adherence will be measured in all participants using an electronic medication monitoring multi-dose pillbox: the Medminder. Enrolment will be followed by a 3-month run-in period, during which group allocation will be concealed and no intervention administered. At the 3-month visit, participants assigned to the intervention group will form an Adherence Support Team including the participant, one or both parents, and a study facilitator who is not a member of the treating team. At the same visit the facilitator will provide standardized adherence education, and will initiate a novel 20-30 min. behavioral intervention, which combines problem-solving skills with implementation intentions (concrete action plans in which an individual specifies, in an if-then contingency format, when, where and how he or she will perform a behavior, with the goal of developing habits). This intervention will focus on addressing adherence barriers identified using the validated Medication Barriers Survey 3 and selected by the participant as important to him or her. Subsequent study visits, at 3-month intervals, will include a briefer versions of the educational component, and review and updating of implementation intentions. In between visits, the facilitator will maintain weekly (for 4 weeks), then monthly contact with participants via short phone or text-message check-ins. In addition, the Medminder will be configured to provide alarm, phone, or text message dose reminders to participants in the intervention group throughout the intervention interval. Control participants will also meet with the facilitator at 3-month intervals, but will simply discuss general health and life issues; they will not receive dose reminders. The primary outcome will be 'taking adherence' - the proportion of prescribed doses that were consumed. Appropriate timing of doses will also be evaluated, as will variability in medication levels (reflecting consistency of medication consumption), and graft outcomes. Level of adherence, patterns of change in adherence, and graft outcomes will be compared between intervention and control groups. Secondary observational analyses of collected study data will identify healthcare systems factors independently associated with adherence.